Lao serial verb constructions and their event representations

• Main Author: Cole, Douglas James
• Other Authors: Davies, William D., 1954-2017
• Format: Others
• Language: English
• Published: University of Iowa 2016
• Subjects: event cognition; lao; serial verbs; syntax; Linguistics
• Online Access: https://ir.uiowa.edu/etd/2197; https://ir.uiowa.edu/cgi/viewcontent.cgi?article=6847&context=etd

This dissertation is an investigation of serial verb constructions in Lao (Tai-Kadai, SVO) and the events that they encode. Serial verb constructions (SVCs), structures where multiple verbs appear in a single clause, raise several important questions for syntactic theory. One issue is how the verbs are related; proposals involving coordination (Payne 1985), subordination (Collins 1997), and adjunction (Hale 1991; Muansuwan 2002) have all been made, while others have made a case for unorthodox double-headed structures (Baker & Stewart 2002; Baker 1989). Additionally, the argument sharing seen in SVCs is seemingly incompatible with proposed constraints on theta-role assignment, such as the Theta-Criterion (Chomsky 1981) or the Biuniqueness Condition (Bresnan 1980). In this thesis I describe new data from the Lao language focusing on two subtypes of SVC that Stewart (1998) calls consequential SVCs (CSVCs) and resultative SVCs (RSVCs). I propose a generative analysis of these structures where an event head licenses a complex VP containing multiple verbs where the object is thematically related to the complex VP rather than the individual predicates. Evidence for the event head comes from a modified version of the explicit segmentation task (Zacks et al. 2001). During the experiment, participants were instructed to divide video clips into events. When participants saw a CSVC before the video, they divided the action sequence depicted by the CSVC into fewer events than when participants saw a coordinated construction before the video. These results suggest that seeing the SVC prompted the participants to group the target sequence of events in the videos together as a larger macro-event, supporting the claim that SVCs encode a single event (contra Foley 2010). These data also support the proposal that events are conceptualized at the clausal level, rather than at the verbal level, which is in line with proposals from Evans (2010), Jackendoff (1991), and Pustejovsky (1991).